Individual differences area: core studies

9.1 · Free notes

Baron-Cohen et al. (2001) The Reading the Mind in the Eyes test

Key points

• Aim: to test the revised Eyes Test as a measure of advanced theory of mind (the ability to infer mental states from the eyes) in adults with Asperger syndrome (AS) or high-functioning autism (HFA) compared with general-population controls.
• Theory of mind (ToM) is the ability to attribute thoughts, feelings and intentions to others. Autism spectrum conditions are linked to a ToM deficit. Adults with AS/HFA often pass basic false-belief tasks, so a more sensitive, advanced test was needed.
• The original (1997) test had only two answer options per item and suffered from floor/ceiling effects. The revision used 36 items, four options per item (one target and three foils) and added a glossary of mental-state words.
• Materials: 36 black-and-white photographs of the eye region only (from magazines), each with four mental-state words, plus a practice item. Each correct choice scores 1, so the maximum score is 36.
• Validity checks: a panel of judges checked that the target word was the best answer for each item. A gender recognition from the eyes task was a control, to show that any deficit was specific to mental-state inference and not to basic perception.
• Sample: Group 1 was 15 men with AS/HFA, recruited through adverts and support groups. Group 2 was 122 general-population adults (adult education classes and public library users), Group 3 was 103 Cambridge University students, and Group 4 was 14 general-population adults matched to Group 1 on IQ.
• Procedure: each person was tested individually, chose the one word that best described what the person in the photograph was thinking or feeling, and could use the glossary. The data were quantitative (scores out of 36).
• Results: the AS/HFA group scored a mean of about 22/36, significantly lower than controls at about 26/36. They did not differ on the gender recognition task.
• Conclusion: the revised Eyes Test is a valid measure of individual differences in ToM, and adults with AS/HFA show a specific impairment in advanced ToM. A low score is not a diagnosis by itself.
• Strengths: a standardised procedure, quantitative data that can be compared statistically, matched controls and large norms. Weaknesses: a small, mostly male AS/HFA sample, static eye photographs (low ecological validity), and culturally specific faces and vocabulary.
• Ethics: participants gave informed consent and could withdraw. Because AS/HFA participants may be vulnerable, researchers must avoid stigma and protect confidentiality.

In more detail

• Worked results: mean scores out of 36 were about 22 for the AS/HFA group, 26 for the general population, 28 for Cambridge students and 31 for the IQ-matched group.
• Sex difference: in the general population, women scored slightly higher than men on average, which fits Baron-Cohen's idea that autistic traits are an extreme form of a more typically male profile.
• AQ link: participants also completed the Autism Spectrum Quotient (AQ), and higher AQ scores went with lower Eyes test scores, so theory of mind varies across everyone, not only in autism.
• Why eyes only: showing only the eye region removes clues from the mouth and the rest of the face, making the task harder and better at picking up subtle differences in adults.